The multilateral trading system and preferential trade agreements: can the negative effects be minimized?

Dieter, Heribert

 Preferential trade agreements pose a big challenge for the
 multilateral trading system. Throughout the first decade of the
 twenty-first century, their number has grown significantly. However,
 these agreements have a range of disadvantages compared with the
 multilateral regime, for example, in trade facilitation and in
 dispute settlement. Whereas it will be difficult to stop the further
 spreading of this wave of preferential agreements, attempts can be
 made to reduce the negative effects of trade agreements that do, by
 definition, discriminate against other countries. In this article, a
 range of potential remedies are discussed, from a moratorium to the
 better enforcement of World Trade Organization rules on preferential
 agreements as well as improved monitoring. KEYWORDS: world trade,
 preferential trade agreements, rules of origin, Article XXIV, World
 Trade Organization.

**********

One of the greatest challenges for the multilateral trading system in the twenty-first century is the rise of preferential trade agreements (PTAs). (1) Both the number and the scope of these agreements are rising rapidly. The collapse of yet another ministerial meeting of the Doha Round, in July 2008 in Geneva, will further fuel the trend toward preferential agreements. Moreover, the global economic and financial crisis has further reduced the willingness of policymakers to support trade liberalization. In many countries, the pendulum is swinging from trade liberalization to economic nationalism. The multilateral trading system is suffering from the economic crisis at the same time as the preferential agreements continue to thrive.

Today, no country or region is abstaining from this trend. Though the European Union (EU), which started with the implementation of a customs union in 1958, has been implementing preferential agreements for many years, countries in other regions (notably, East Asia) did not contribute to the rising number of preferential agreements before the year 2000. In recent years, however, Asian economies, including that of China, have contributed to this dangerous trend.

Due to the collapse of the July 2008 ministerial meeting in Geneva, there will be increased demand for PTAs. The arguments we have heard in recent years will be repeated. A lack of progress in the multilateral arena will be given as the prime reason for preferential agreements and, of course, this argument is now more powerful than before. With no agreement on the Doha Development Round in sight, the global economy is poised for a new round of preferential agreements.

The failure of the 2008 Geneva talks has some parallels with the London Economic Conference of June 1933. Although the purpose of the 1933 conference was not to achieve an agreement on trade, but rather on financial affairs, the willingness to sacrifice multilateral cooperation in favor of narrow national economic goals is a striking parallel. Both then and now, the United States has been suffering from self-inflicted economic turmoil and unwilling to make sufficient concessions to its economic partners. (2) Following the failed London conference, the global economy became less integrated and preferential agreements were concluded all over the world, leading to fragmentation rather than integration. The 1930s became a decade of noncooperation, where "multilateralism was replaced by bilateralism, non-discrimination by discrimination, free trade by comprehensive protection, freedom for capital flows by exchange controls and free movement of labour by rigorous restrictions." (3)

Of course, parallels with the 1930s should not be overdrawn. Nevertheless, policymakers should be aware of the risks that PTAs pose for international economic relations. (4) By definition, preferential agreements exclude countries. PTAs are liberalizing and countries do make concessions in them, but these are limited to the signatories and not extended to others. Thus, preferential agreements discriminate. This, of course, weakens nondiscrimination, which was the central pillar of the postwar trade regime. Nondiscrimination was intentionally the core norm of the General Agreement on Tariffs and Trade (GATT), embodied in Article I. The idea behind this clause was not purely, perhaps not even primarily, economic. Nondiscrimination was considered to contribute to the stability of international relations. Increasingly, policymakers seem to forget that international relations cannot flourish in an atmosphere of discrimination and exclusion.

Against these historic lessons, the first best solution would be to eliminate preferential agreements altogether. In a world without preferential agreements, countries could either have unilateral restrictions on trade or agree in multilateral forums on liberalization measures. Today, however, this is not a plausible proposal. All countries, or at least all World Trade Organization (WTO) member countries, would have to agree on the prohibition of PTAs. This appears to be pure wishful thinking. In an era of noncooperation that is characterized by the increasing unwillingness or inability of the major players to reach consensus on vital issues, how can one realistically expect policymakers to forgo a policy instrument that is so attractive to them? (5) Thus, the challenge today is to implement measures that minimize the negative consequences of PTAs, and canvassing such measures is one of the purposes of this article.

Several questions have to be asked about the causes and the consequences of the trend toward PTAs. First, what are the motives for policymakers to push preferential agreements rather than regulation in a multilateral context? Second, what are the negative consequences of preferential agreements, in particular, for developing countries? Third, how can these trade agreements be modified so that their negative effects will be minimized? These three questions will be discussed in turn.

The Rise of Preferential Agreements and Their Variance in Form

In the 1950s and 1960s, the number of PTAs notified to the GATT was small, resulting in a total of less than ten agreements at the end of the 1960s. In recent years the number of notifications had risen to almost 400 by the end of 2007. (6)

In contrast to previous decades, virtually all players in global trade in the early twenty-first century are engaged in preferential agreements. In particular, the United States, the EU, Japan, and China have been pushing these deviations from nondiscrimination. The United States now seems to offer PTAs to those countries that are willing to accept the (market opening) US template for preferential agreements, which puts particular emphasis on intellectual property rights and market opening in financial services, and to those countries whose foreign and security policy is in accord with that of the United States. (7) The EU has joined this trend and has recently given up its moratorium on new agreements. In Asia, both China and Japan have been actively pursuing PTAs since the turn of this century.

One of the great paradoxes of today is the fact that the member countries of the WTO demonstrate a wide gap between their rhetoric, which continues to be multilateral, and their policy actions, which favor preferential agreements. The existence of an alternative to the multilateral system, even if that alternative is a flawed one, may have been the decisive factor that has hindered a successful conclusion of the Doha Round thus far and conceivably forever. But what are the motives of policymakers for negotiating and concluding preferential agreements?

* The length of negotiations in the current round causes difficulties for democratically elected governments, which tend to operate within shorter electoral cycles of three to five years. Preferential agreements can be tailored to fit into such time frames.

* Bilateral agreements seem to provide immediate, reciprocal gains for each country's own economy. This perception allows governments to work with coalitions supporting liberalization, which are instrumental in overcoming internal opposition to an agreement. Reciprocity is a useful political tool and is more easily promoted to domestic constituencies than some arcane economic theory about the benefits of unilateral or multilateral trade liberalization.

* Transnational corporations are increasingly pushing the implementation of preferential agreements. Two distinct motivations can be identified. First, some companies are pursuing an offensive strategy, seeking the opening of hitherto closed markets (e.g., in financial services). Second, other companies claim defensive motives, suggesting that the implementation of free-trade agreements by third countries puts them at a competitive disadvantage. (8)

* The vanity of politicians and trade negotiators contributes to the current trend. Negotiators and politicians do not receive much praise, if any, for successfully concluding a multilateral agreement, but preferential agreements seem to prompt a more positive media response and are an opportunity to enjoy television coverage, the "CNN effect." Bilateral agreements allow the leaders of not-so-great powers to have their fifteen minutes of fame, especially, in a bilateral deal with a major global or regional power. The prospect of media coverage may encourage the engagement in bilateral negotiations. The CNN effect can be an important visible expression of state sovereignty and it can boost regime authority. In addition, some agreements are concluded because of diplomatic pressures (i.e., the willingness of foreign policy elites to express positive political relations with a trade agreement).

Of course, this short list of reasons for the emergence of the strong trend for PTAs is not comprehensive. (9) However, it demonstrates that the motives for implementing preferential agreements are embedded in the political systems of the WTO member states. This pattern is unlikely to change in the short or medium term. Rather, the global economic crisis has strengthened protectionist thinking in many economies, both developed and developing. Although the conclusion of the Doha Round is now quite unlikely, countries may continue to implement preferential agreements. In an environment characterized by fear and risk aversion, policymakers may find it considerably easier to promote bilateral agreements, which demonstrate a more obvious utility to an economy than the more idealized concept of the multilateral trade regime.

In addition to the variance that we can observe with regard to the partners in PTAs, their scope does show great diversity. In recent years, some agreements have gone well beyond trade in goods and services, and have addressed policies on mutual recognition of standards, competition policy, and the movement of persons as well as investment and cooperation agreements. (10) Thus, we observe a range of agreements, some of which go far beyond the regulation of trade. Of course, the interest of countries to pursue so-called WTO-plus agreements also reflects the problems of agreeing on new issues of economic governance, both within the WTO and in other multilateral organizations.

The Disadvantages of Preferential Trade Agreements

In principle, there are two schools of thought on preferential agreements. The first one argues that these agreements are contributing to deeper global integration. The second school of thought questions the utility of these agreements and considers them to be dangerous for international (economic) relations.

Obviously, one could argue that as long as trade liberalization occurs, it does not matter whether it is at a bilateral, regional, or multilateral level. In theory, preferential agreements can have significant benefits (e.g., promoting technology and knowledge transfers, domestic reforms, productivity gains, and improved developmental prospects). By contrast, the critics of preferential agreements emphasize the negative effects, including the distortion in trade patterns between "insiders" and "outsiders" which undermines the welfare gains arising from expanded trade.

The most critical and widely recognized issue with regard to preferential agreements is the erosion of GATT's fundamental nondiscrimination principle. The three key exceptions to this principle are found in Article XXIV of GATT, permitting PTAs; the Enabling Clause of 1979, permitting preferential treatment of goods from developing states; and Article V of the General Agreement on Trade in Services (GATS), which is similar in intent to Article XXIV. Today, the increased use of these exceptions is reducing the most-favored-nation clause to the exception rather than the rule.

A major disadvantage of all preferential agreements is the need to establish the "nationality" of a product. In an entirely open world economy with no restrictions on the flow of goods, rules of origin would not matter. Today, however, the origin of a product does matter in preferential agreements. All PTAs require rules of origin to establish the nationality of a product, given that participating countries continue to have diverging external tariffs. Because only goods produced within the territory of the agreement qualify for duty-free trade between the partners of the agreement, there have to be procedures that differentiate between goods produced within and goods from the rest of the world. The preferential system becomes complicated and expensive. The administrative burden of issuing certificates of origin is, of course, most problematic for those countries that have limited resources (i.e., developing countries).

 What makes RoO (rules of origin) particularly relevant is that they
 are not a neutral instrument: given that RoO can serve as an
 effective means to deter transhipment, they can tempt political
 economy uses well beyond trade deflection. Indeed, RoO are widely
 described as a trade policy instrument that can work to offset the
 benefits of tariff liberalization.'' (11)

The negative consequences of rules of origin rise with the number of agreements implemented. Furthermore, with multiple agreements, companies are faced with diverging rules and procedures, which add to the costs of generating certificates of origin. The operational costs of meeting diverse requirements in different countries pose a major challenge, especially for small- and medium-sized enterprises. Preferential trade agreements with these outcomes are clearly a second-best option compared to a multilateral agreement at the WTO with uniform rules applicable to all members.

Some observers have suggested that, over time, the negative consequences of preferential agreements will become so obvious that countries will stop implementing them and will return to the multilateral forum. This may be the case, but waiting for such a learning process by policymakers in the WTO membership seems to be too little. Furthermore, the negative consequences of preferential agreements have been known for decades, and yet the availability of this knowledge has led only to some caution with regard to new preferential agreements (e.g. the unilateral moratorium EU trade commissioner Pascal Lamy implemented after 1999).

We continue to know relatively little about the degree to which preferences are utilized by companies. Available evidence is limited. A study by the Canadian government suggests that, even in the North American Free Trade Agreement (NAFTA), one of the older preferential agreements, only about 30 percent of preferences are utilized because of both the low level of existing tariffs and the cost of establishing origin. (12) In other agreements, such as the Association of Southeast Asian Nations (ASEAN), the utilization rate of under 10 percent is even lower. (13) For many companies, paying the tariff is more attractive than incurring the cost of complying with the bureaucratic procedures for establishing origin. (14)

But there are additional disadvantages of preferential agreements. In particular, transferring dispute settlement to the bilateral level can be problematic, especially, for the weaker partner in a PTA. In many bilateral schemes, there is an option--either bilateral dispute settlement or multilateral dispute settlement. It is obvious that the bilateral route offers many possibilities for the more powerful partners to promote their case. Hierarchy and power--never fully absent in international trade--have a more prominent role in PTAs than in the multilateral regime. The existence of an alternative to the WTO dispute settlement mechanism provides the more powerful countries with an additional choice but, for weaker countries, this can be a drawback. In some cases, bilateral agreements have no provisions for dispute settlement, leaving the less powerful trade partner disempowered in the event of a conflict over a trade measure. Moreover, bilateral trade agreements are not supported by the Advisory Centre for WTO Law that can provide legal advice to developing countries and Least Developed Countries involved in a dispute through the WTO. (15)

How Can the Negative Effects of Preferential Agreements Be Minimized?

It is widely recognized that preferential agreements are a suboptimal way to organize world trade. The discrimination that is endemic in these arrangements is economically inefficient. Moreover, the deepening of international production networks, through the sourcing of inputs from a broad range of countries, raises the costs of preferential agreements. (16) Now that the WTO membership has been unable to reach an agreement on modalities that would have significantly progressed the Doha Round in July 2008, WTO members face a choice. They can either continue to observe the spread of preferential agreements without changing the rules, or they can address the problem by trying to minimize their negative effects.

At a quite general level, one way of reducing the attractiveness of preferential agreements would be to further reduce most-favored-nation (MFN) tariffs. (17) If these tariffs are below 5 percent, the costs associated with establishing origin in a preferential agreement are equal to or even greater than the tariff, so the interest in preferential agreements would probably fade away quickly. However, the uneven structure of tariff lines has to be considered. Although many tariff lines are already quite low today, there continue to be high tariffs on specific products in both developing and developed economies. (18)

Given the paralysis in multilateral negotiations and the limited willingness of policymakers to unilaterally liberalize much further at this point, the question is whether a political initiative by important countries could send a signal against the further spreading of preferential agreements. The Warwick Commission has suggested that the major industrialized countries as well as large developing countries should refrain from establishing PTAs with each other. (19) This recommendation was made with the worst-case scenario in mind in which the EU, United States, and major Asian economies negotiated PTAs among themselves. Were such a situation to eventuate, the WTO would face the prospect of irrelevance. Although not likely today, we have seen some developments in that direction, including South Korea's bilateral negotiations with the United States and the EU, and negotiations for a transatlantic marketplace between the United States and the EU. (20)

Beyond the recommendations of the Warwick Commission, a moratorium on new preferential agreements is conceivable in Organisation for Economic Co-operation and Development (OECD) countries. For such an agreement, two ingredients would be essential. First, an alternative to preferential agreements would be necessary. So-called critical mass initiatives, in which subsets of WTO members make agreements whose benefits are then extended to the entire membership, would constitute such an alternative. Second, the countries supporting such an initiative would have had to come to terms with their policies toward globalization. Today, we observe a widespread reluctance to underwrite the concept of further international division of labor. (21) Thus, governments in OECD countries may see little benefit in initiatives that risk electoral backlashes, even though the direct effect of a moratorium on preferential agreements would be to safeguard an established and valuable multilateral regime.

Quite promising are reforms that would directly address one of the most crucial problems of today's multilateral trade negotiations. Up until today, an unhealthy dichotomy has emerged. Because the Doha Round is a "single undertaking" (i.e., nothing is agreed until everything is agreed), countries are using preferential agreements to achieve some progress in specific areas. The answer to that dilemma could be the introduction of critical mass initiatives under the roof of the WTO. Inter alia, the Warwick Commission has suggested this approach. (22) In essence, countries willing to agree on certain specific policies could go ahead and implement policies without needing the consent of all WTO member countries. In effect, preferential agreements would cease to be the only alternative to the multilateral regime. Although a critical mass approach does have disadvantages (see the contribution of Peter Gallagher and Andrew Stoler in this special issue), on balance this avenue is superior to the continuation of an uncontrolled and, given their popularity, probably uncontrollable spreading of preferential agreements.

One way of exercising greater control of preferential agreements is to strengthen regulation within the WTO. Preferential agreements are permitted under Article XXIV of GATT, which permits preferential agreements subject to certain conditions. However, the disciplines imposed by Article XXIV lack definitional clarity and have never been enforced. The key question is whether there are ways to discipline their use. In theory, preferential agreements could be reviewed under two different mechanisms. First, member countries must notify the WTO about agreements, which are subsequently discussed in the Committee on Regional Trade Agreements. In practice, this track has been of limited utility. All WTO member countries can participate in the committee, including the states participating in the preferential agreement. Although, in theory at least, the committee could by consensus deem a notified agreement to be WTO inconsistent, as a practical matter the committee probably never would reach such a consensus.

Second, in principle, a WTO member could challenge the WTO consistency of a preferential agreement through the dispute settlement proceedings. Hitherto, member countries have refrained from pursuing this avenue. The legality of preferential agreements has not been challenged. A plausible explanation is that virtually all member countries have been implementing preferential agreements and could therefore face a challenge to their preferential agreements if they initiate a legal challenge against other countries. Consider, for example, a potential legal challenge of the legality of US trade agreements by the EU. The dispute settlement mechanism of the WTO would either be blocked for years or the legitimacy of the dispute settlement might be questioned by one of the major players.

Moreover, given the uncertainty surrounding the precise legal frontiers of the line between lawful and unlawful preferential agreements--and the widespread acceptance of them on a de facto basis by WTO members--dispute panels should probably avoid making a definitive decision on the WTO consistency of any particular agreement. Political issues have to be solved by policymakers, not the juridical system.

However, there is room for improvement within the sphere of activity of the WTO. The Committee on Regional Trade Agreements, which appears to be at best lacking courage and at worst moribund, should be given a clear and strong mandate for the improvement of the supervision of preferential agreements. One significant development in the WTO was the adoption in December 2006, on a provisional basis, of a Transparency Mechanism for PTAs. The Transparency Mechanism requires members to notify of a new PTA as soon as possible, and no later than immediately following its ratification. The WTO secretariat then prepares a factual presentation on the detail of the new PTA, in consultation with the signatories involved. This is an important, albeit limited, development; as in other areas of regulation, increasing transparency is only a first step. Even complete information will not alter the structural problems that are related to preferential agreements. To achieve this, bolder steps would be necessary.

The inherently discriminatory and puzzling system of PTAs questions the WTO's ability to manage the increasingly complex system of trade governance. If the WTO and its members are unwilling to stop the glut of preferential agreements, one potential avenue for reducing the negative effects of this trend would be to transform the WTO into a supervising agency. In such a scheme, the WTO would monitor, evaluate, and, where necessary, sanction PTAs. Member countries of the WTO would continue to be free to implement preferential agreements, but the criteria for permissible agreements would be sharp and coherent. Needless to say, such a regime would require a reasonably strong instrument for sanctioning unacceptable agreements. The simplest, and yet most forceful, one would be to allow the WTO to open the preferences granted in the preferential agreements to all other WTO member countries. Thus, if the WTO supervising body concludes that one element of a particular preferential agreement violates the WTO rules, other member countries would be permitted to retaliate by being permitted to take advantage of the agreement's preferences.

The WTO's evolution into a supervisory body--overseeing preferential agreements, developing a transparency mechanism, and sanctioning nonconforming agreements--would constitute an effective response to the recent proliferation of preferential agreements. Nonetheless, this evolution would require two substantial policy shifts. First, member countries would have to agree on what constitutes an acceptable preferential agreement and what does not. In essence, the currently vague regulations of Article XXIV would have to be expanded and more explicit formulas would have to be found. For example, the term "substantially all the trade" would have to be made specific. Does that mean 80, 90, or 99 percent of all existing trade? Or would the regulation refer to all potential trade (i.e., including areas that countries considered to be unsuitable for increased competition from foreign companies)? Second, member countries of the WTO would have to give the organization a widened mandate and permit it to interfere in their foreign economic policy. Although this has been the case in the past, giving the WTO explicit jurisdiction over preferential agreements would be an additional step.

In order to know which agreements have less negative effects than others, it would be necessary to develop best-practice guidelines for preferential agreements. Member countries would have to negotiate these. (23) By developing such a catalog, the effects of any preferential agreement would be much more transparent to both those involved in negotiations and to third parties.

However, the question is whether this is a plausible proposal. First, Article XXIV of GATT does contain some guidelines (e.g., the provision that "substantially all the trade" ought to be covered by the agreement). Yet member countries have frequently violated the principle, for example, by excluding agricultural products more or less completely from preferential agreements. (24) Political pressures in member countries, in particular the powerful agricultural lobbies, will be an obstacle for the implementation of a rigid implementation of standardized regulations on openness. Second, some countries have used preferential agreements to advance the interests of domestic industries. A prominent example is the financial sector of the United States, which managed to secure the prohibition of restrictions on capital flows in some agreements (e.g., in those with Singapore and Chile). The question is whether these interest groups could be pacified with a best-practice agreement that would restrict the remit of preferential agreements to narrowly defined trade-related issues (of course, restrictions on capital flows are not).

Some of the proposals made in the discussion on preferential agreements are ignoring the fact that any binding agreement on best practice would not be easily accepted in democratic societies. Baldwin and Thornton suggest that "nations would declare themselves and all of their RTAs [regional trade agreements] as subject to this anchorage-building discipline. The benefit of self-declaration would be the signal it provided to potential investors that the nation was permanently committed to pro-market reforms." (25) Such expectations, however, are politically naive and do underestimate the level of resistance that so-called promarket reforms have been generating in recent years. Whereas the Baldwin-Thornton approach may work in autocratic societies, it most probably will not work elsewhere. Consider, for example, the opposition in EU countries against its failed constitution. In the debate, many opponents claimed that the economic policies enshrined in the constitution where too market friendly and did not give sufficient consideration to social issues. In developing countries, there has been enormous, lasting resistance against the Washington Consensus, another variation of seemingly straightforward principles. John Williamson's description of policies that were considered to be essential for economic reform could have been considered to be a set of best practices, yet the academic and political reaction against them was over-whelming. (26)

Another example of the failure to establish a set of trade-related best practices are the so-called Singapore issues, which refer to transparency in government procurement, trade facilitation, trade and investment, and trade and competition. Opposition to this set of best practices was opposed by many developing countries, most notably, during the failed WTO ministerial in Cancun in 2003.

This list is by no means complete, but it shows that the establishment of a consensus on economic policies at a global level is an extremely complex endeavor. (27) In an era in which the process of globalization is contested both in developing and developed countries, the development of a set of enforceable rules on PTAs will be no less difficult than previous attempts to establish best practices.

Nevertheless, there are opportunities for reducing the negative effects of preferential agreements that appear to be more easily achievable. First, regions could decide to follow the example of the European countries and establish zones that accept so-called diagonal cumulation of origin. In effect, the region from which inputs can be sourced can be dramatically increased. The pan-European cumulation system (PECS), introduced in 1997, has eased trade within that system significantly: any good that is deemed as originating in one country of the PECS zone has to be granted originating status in every other country in the zone. (28) Developing, for example, a pan-Asian and a pan-American cumulation scheme would ease trade discrimination at least within those regions. Of course, the unwelcome side effect would be that the emergence of three large trading blocs would be facilitated.

Second, preferential rules of origin could be harmonized. Ideally, this would result in their standardization. If there could be one set of criteria--uniform for all preferential trade agreements--the negative effects would be significantly reduced. However, some observers argue that a complete harmonization of rules of origin would be politically difficult because it would be "unpalatable to producers around the world." (29) Although it is true that striking a bargain on rules of origin would be difficult, it does not appear to be an insurmountable hurdle. If the beneficiaries of a liberalization measure are sufficiently powerful to withstand opposition, there is no need to compensate the losers. If they are not, compensation is necessary. Of course, mobilizing political support for such a highly technical and complex area of regulation will not be easy.

Third, dispute settlement could be made exclusive to the WTO. Such a step would require that all preferential agreements would have to contain a clause that grants the WTO exclusive jurisdiction on its disputes. Of course, some larger countries or entities would probably oppose such a scheme, in particular the EU, which likely would have no intention to give up jurisdiction over issues related to the single market. A potential solution could be that the WTO would retain the jurisdiction for all free-trade agreements, but permit the development of dispute settlement mechanisms for customs unions and common markets.

Another potential avenue for reducing the negative effects of preferential agreements would involve multilateralizing commitments that countries make in preferential agreements. In areas such as procurement, services, investment, and competition policy, many countries (including developing nations) have been willing to make commitments in preferential agreements, but have resisted attempts to negotiate these matters at the multilateral level. (30) The WTO could launch initiatives to multilateralize those commitments, and such an endeavor would, if successful, reduce the negative effects of preferential agreements. (31) The risk, of course, is that the WTO could itself come under attack from member countries for pushing them into a certain direction. If the organization itself were to push such initiatives, the WTO could lose credibility and legitimacy in some member countries. The potential losses would probably outweigh the potential gains.

Conclusion

On balance, PTAs do have very few advantages. They are inferior to regulating trade at the multilateral level, and not many of them support deeper integration as do some of the larger regional agreements (e.g., the EU, NAFTA, and ASEAN). Preferential trade agreements are a third-best solution for regulating international trade. They are suboptimal with regard to economic efficiency, and they are imbalanced because they disadvantage the poorer players and systemically strengthen the more developed players. Despite these obvious drawbacks, a fundamental revision of countries' trade policies appears to be unrealistic. In the early twenty-first century, the willingness of nations to support the existing institutions of economic governance appears to be limited and, consequently, preferential agreements are here to stay.

However, accepting the inevitability of a further spreading of PTAs does not mean that their negative effects cannot be reduced. Among the options discussed in this article, reforming the negotiating process with the WTO is the most promising avenue. The introduction of critical mass initiatives under the supervision of the WTO would enable members to form coalitions of like-minded countries that could advance issues of common concern without having to opt out of the multilateral regime.

In addition, member countries could provide the WTO with a clear mandate for monitoring and supervising PTAs, including the application of sanctions for agreements that violate Article XXIV. The harmonization of preferential rules of origin would be a further step to reduce the negative effect of PTAs.

By contrast, the development of an agreed set of best practices for preferential agreements appears to be difficult. The definition of what constitutes best practice inevitably varies according to the level of development of a country as well as the political preferences of societies. Past experience has demonstrated how controversial best practices can be.

However, the changing geopolitical context appears to be more important than the details of trade agreements. As discussed in the introduction, we may currently witness a departure from multilateral governance not entirely different from the 1930s. The United States may decide that the disadvantages of globalization outweigh the benefits and put more emphasis on preferential agreements in line with other OECD countries. Early statements from me Obama administration are not encouraging. In March 2009, the United States signaled a tough position on Doha and expressed the belief that the Doha Round was "tilted against the US." (32) In his campaign for the presidency, Barack Obama expressed reservations on the utility of a liberal regime, raising concerns for instance about the loss of jobs due to offshoring. He even criticized NAFTA for not considering the needs of American workers, (33)

The combination of a severe and lasting economic crisis, increasing hostility toward globalization in the United States, and a lack of supporters of the multilateral trade regime elsewhere may result in a fatal blow to the World Trade Organization. The return to the world of the 1930s, not long ago deemed unlikely, is a considerable risk today. Preferential trade agreements would play a prominent role in a world characterized by discrimination.

Effectively, the global economy seems to be returning to a regime where goods originating from befriended countries have easier access to a national market than others. There is discrimination between friends and foes. The postwar trading regime had the explicit goal of nondiscrimination, and today's policymakers are sacrificing this philosophy on the altar of seemingly quick, but asymmetrical and eventually unsustainable, economic gains.

Notes

Heribert Dieter is a senior fellow at the German Institute for International and Security Affairs, Berlin, and teaches at die University of Potsdam.

(1.) Throughout this article, I will use the term preferential agreements for all types of trade agreements that exclude other countries (i.e., bilateral and plurilateral free trade agreements and customs unions).

(2.) President Franklin D. Roosevelt's rejection to contribute to the stabilization of exchange rates was the final nail in the coffin of the gold standard; see Harold James, International Monetary Cooperation Since Bretton Woods (New York: Oxford University Press, 1999), p. 25.

(3.) Martin Wolf, "Is Globalization in Danger?" The World Economy 26, no. 4 (April 2003): 393-411.

(4.) For a discussion on the tensions between regional and multilateral regulation, see Peter Katzenstein, "Regionalism in Comparative Perspective," Cooperation and Conflict 31, no. 2 (June 1996): 123-159.

(5.) For a discussion of the changing geopolitical context, see Heribert Dieter and Richard Higgott, Between Crisis and Reform: Governing the Global Economy in an Era of Non-cooperation (forthcoming), chap. 1.

(6.) Some 300 of them are trade agreements covering trade in goods and are notified under Article XXIV while 58 cover trade in services and are notified under Article V of GATT, and 22 are notified under the Enabling Clause of 1979 (i.e., agreements between developing countries).

(7.) Countries actively supporting the invasion of Iraq (e.g., Australia) were given the "privilege" of a free trade agreement (Bob Zoellick, US trade representative, New Statesman, 23 June 2003, p. 17) while a country like New Zealand, which refused to join the invasion forces, did not get such an agreement. (For a discussion of the securitization of US trade policy, see Heribert Dieter and Richard Higgott, "Linking Trade and Security in Asia: Has Washington's Policy Worked?" Aussenwirtschaft 62 [2007]: 151-174.)

(8.) One example is the European car industry, which pushes preferential agreements between the EU and ASEAN countries, citing these countries' preferential agreements with Japan as their prime concern. German industry has been lobbying for preferential agreements with Asian economies for a few years, typically citing the preferential agreements of other large economies and arguing that abstaining from the trend would do harm to their business interests. See, for example, the press release of the Federation of German Industries in April 2007, which welcomes the push for bilateral trade agreements (Bundesverband de deutschen Industrie, or BDI), available at www.bdi-online.de/download/PMbilateraleFreihandelsabkommen.pdf.

(9.) For a more detailed discussion see, for example, Richard Baldwin, "Multilateralising Regionalism: Spaghetti Bowls as Building Blocs on the Path to Global Free Trade," The World Economy 29, no. 11 (2006): 1451-1518; Ross Garnaut and David Vines, Sorting out the Spaghetti. On Reducing the Damage from the Proliferation of Discriminatory Regional Free Trade Areas (New York: University of Columbia Business School, January 2006), available at http://www0.gsb.columbia.edu/ipd/pub/Vines_Reaction.pdf; John Ravenhill, "The New Bilateralism in the Asia Pacific," Third World Quarterly 24, no. 2 (1999): 299-317.

(10.) John Whalley, "Recent Regional Agreements: Why So Many, Why So Much Variance in Form, Why Coming So Fast, and Where Are They Headed?" The World Economy 28, no. 30 (2008): 518.

(11.) Antoni Estevadeordal, Jeremy Harris, and Kati Suominen, "Multilateralising Preferential Rules of Origin Around the World," paper presented at the conference "Multilateralising Regionalism," Geneva, September 2007, available at www.wto.org/english/tratop_e/region_e/con_sep07_e/estevadeordal_harris_suominen_e.pdf.

For a detailed discussion of the effects of rules of origin, see, for example, Anne O. Krueger, "Free Trade Agreements as Protectionist Devices: Rules of Origin," National Bureau of Economic Research Working Paper No. 4352, Washington, DC, 1993; Kala Krishna and Anne O. Krueger, "Implementing Free Trade Areas: Rules of Origin and Hidden Protection," National Bureau of Economic Research Working Paper No. 4983, Washington, DC, 1995; Antoni Estevadeordal and Kati Suominen, "Rules of Origin: A World Map and Trade Effects," paper presented at the conference "The Origin of Goods: A Conceptual and Empirical Assessment of Origin in PTAs," Paris, May 2003, available at www.inra.fr/Intemet/Departements/ESR/UR/lea/actualites/ROO2003/articles/estevadeordal.pdf; Heribert Dieter, "Praferenzielle Ursprungsregeln in Freihandelszonen: Hemmnisse fiir den internationalen Handel?" Aussenwirtschaft: Schweizerische Zeit-schrift fur Internationale Wirtschaftsbeziehungen 59, no. 3 (2004): 273-303; Estevadeordal, Harris, and Suominen, "Multilateralising Preferential Rules of Origin Around the World."

(12.) Whalley, "Recent Regional Agreements," p. 522.

(13.) Baldwin, "Multilateralising Regionalism," p. 1488.

(14.) Richard Baldwin and Phil Thornton, Multilateralising Regionalism: Ideas for a WTO Action Plan on Regionalism (London: Centre for Economic Policy Research, 2008), p. 14.

(15.) Christina L. Davis, "Do WTO Rules Create a Level Playing Field? Lessons from the Experience of Peru and Vietnam," in John Odell, ed., Developing Countries and the Trade Negotiation Process (Cambridge: Cambridge University Press, 2006), p. 239.

(16.) Baldwin and Thornton, Multilateralising Regionalism, p. ix.

(17.) Richard Baldwin, Simon Evenett, and Patrick Low, "Beyond Tariffs: Multilateralising Deeper RTA Commitments," paper presented at the conference "Multilateralising Regionalism," Geneva, September 2007, p. 4, available at www.wto.org/english/tratop_e/region_e/con_sep07_e/baldwin_evenett_low_e.pdf.

(18.) A few examples from the automotive sector illustrate this. For example, the United States still charges 25 percent duty on so-called light trucks, the currently unpopular commercial vehicles. A relatively advanced country like Thailand has an import duty of 80 percent on larger cars, and the EU continues to have an import duty of 10 percent on all vehicles.

(19.) Warwick Commission, The Multilateral Trade Regime: Which Way Forward? University of Warwick, December 2007, p. 53, available at www2.warwick.ac.uk/research/warwickcommission/report/uw_warcomm_tradereport_07.pdf.

(20.) For a discussion of the risks inherent to a transatlantic trade agreement, see Rolf Langhammer, "Why a Market Place Must Not Discriminate: The Case Against a US-EU Free Trade Agreement," Kiel Working Paper No. 1407, March 2008.

(21.) For a discussion of global economic governance in the twenty-first century, see Dieter and Higgott, Between Crisis and Reform.

(22.) See Warwick Commission, The Multilateral Trade Regime, pp. 30-31.

(23.) Baldwin and Thornton, Multilateralising Regionalism, p. 38.

(24.) For example, the preferential agreement between Japan and Thailand, concluded in 2005. excludes rice, wheat, beef, dairy products, and fish. Of course, these are highly protected sectors of the Japanese economy; thus, gains from liberalizing would be greatest there.

(25.) Baldwin and Thornton, Multilateralising Regionalism, p. xi.

(26.) John Williamson, What Washington Means by Policy Reform (1990; reprint, Washington, DC: Institute for International Economics, November 2002), available at www.iie.com/publications/papers/williamson1102-2.htm. Some of the policies (e.g., the proposal for the privatization of state enterprises) were controversial but, by and large, the recommendations were motherhood and apple pie. Capital account liberalization, an important element of the financial crisis of the 1990s, was not included in the Washington Consensus.

(27.) One could add the failed Multilateral Agreement on Investment (MAI), in effect abandoned in 1998, as well as the fiasco that the International Monetary Fund experienced with its contingent credit lines, which would have required countries to accept a set of best practices.

(28.) Baldwin and Thornton, Multilateralising Regionalism, p. 32.

(29.) Estevadeordal, Harris, and Suominen, "Multilateralising Preferential Rules of Origin Around the World," p. 44.

(30.) Baldwin, Evenett, and Lo, "Beyond Tariffs," p. 35.

(31.) For a discussion of potential initiatives, see Baldwin, Evenett, and Low, "Beyond Tariffs," pp. 36-38. For a comprehensive debate, see Richard Baldwin and Patrick Low, eds., Multilateralizing Regionalism (Cambridge; Cambridge University Press, 2009).

(32.) Alan Beattie, "Obama Signals Tough Stance on Doha," Financial Times, 2 March 2009, available at www.ft.com/cms/s/0/cf8757b4-075b-11de-9294-000077b07658.html.

(33.) Barack Obama on NAFTA, available at Greg Mankiw's website at http://gregmankiw.blogspot.com/(accessed 21 February 2008).